WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies

Page 1 of 10
Research, Education, Technical Support & Information
ABN 69 003 696 526
Unit 3, Suite 2, 9 Parramatta Road, Lidcombe, 2141. PO Box 6165, Silverwater, NSW 1811 Australia
Phone: (02) 9748 4443 Fax: (02) 9748 2858 E-mail info@wtia.com.au URL http://www.wtia.com.au
GUIDANCE NOTE 6
POST WELD HEAT TREATMENT OF WELDED STRUCTURES
INTRODUCTION
During the fabrication process, welding is
the most commonly used method of joining
items together. The welding process
generally involves melting and subsequent
cooling, and the result of this thermal cycle
is distortion if the welded item is free to
move, or residual stress if the item is
securely held. There comes a point when the
amount of residual stress can create potential
problems, either immediately or during the
life of the welded structure, and it needs to
be reduced or removed. Post weld heat
treatment is the most widely used form of
stress relieving on completion of fabrication
of welded structures. The principle is that as
the temperature is raised, the yield stress and
the elastic modulus of the material fall. A
point is reached when the yield stress no
longer supports the residual stresses and
some localised plastic deformation occurs.
The objective of this Guidance Note is to
describe the various post weld heat
treatment techniques, when it has to be used
and the circumstances where it can
legitimately be avoided. The Note is also
intended to assist the Fabrication Industry in
providing information on where to look for
guidance on post weld heat treatment
heating rates, cooling rates, holding
temperatures, holding times, thermocouple
placement, thermocouple attachment, and
qualification of welding procedures. It is
intended for use by Welding Supervisors,
Welding Inspectors, fabrication personnel
and Engineers involved in both the
fabrication industry and typical
Owner/Operator industries.
SCOPE
Post weld heat treatment in the context
described here refers to the process of
reheating a weld to below the lower
transformation temperature at a controlled
rate, holding for a specific time and cooling
at a controlled rate. No consideration has
been given to normalising of welds in
carbon-manganese and low alloy steels, to
solution annealing of stainless steels or to
any precipitation hardening treatments to
other alloy materials. Therefore, almost by
exclusion, the materials involved will be C-
Manganese steels, C-Mo, 1CrMo,
2Cr1Mo, 5Cr1Mo, 9Cr1Mo and
12Cr1MoV steels.
WHY THE NEED FOR POST WELD
HEAT TREATMENT?
Residual Stress
The development of residual stresses
approaching or even exceeding the yield
stress is possible when welding thick
sections. For certain industry sectors, eg.
Petrochemical, Chemical, Oil and Gas, etc.
the existence of residual stress of this
magnitude is completely unacceptable. With
pressure equipment operating at 200
0
C and
below a variety of stress corrosion cracking
mechanisms under the general term
environmental cracking become prevalent.
There is also the problem of fatigue to be
considered and the effect that residual
tensile stresses have in that regard.
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 2 of 10
Other industries such as Power Generation
have different considerations, and the stress
developed due to thermal expansion of
pipework can take on a far greater
significance than residual stress due to
welding. It is interesting to note that it can
be shown (1) by the use of the Larson-Miller
parameter that residual stresses in non-post
weld heat treated welds are almost fully
relaxed after approximately 20,000 hours at
a service temperature of 540
0
C.
Tempering Effect
Post weld heat treatment will generally
result in a modification of the microstructure
of both the weld metal and heat affected
zone. With the exception of the 9Cr1Mo and
12Cr1MoV materials, the microstructure of
all other materials should contain a mixture
of ferrite and iron or alloy carbide. The
effect of short-term (1 to 2 hours) post weld
heat treatment on the carbide is generally
beneficial, whereas longer times result in a
reduction in toughness due to spheroidising
effects. The normal microstructure for the
parent, weld and HAZ for the 9Cr1Mo and
12Cr1MoV materials is martensite, and post
weld heat treatment is absolutely essential in
these materials to temper the martensite
phase.
Effect on Mechanical Properties
As a series of very general statements, the
following are the consequences of post weld
heat treatment compared with the as-welded
condition:
Yield strength is decreased slightly, the
effect falling off fairly rapidly with time.
The tensile strength is decreased.
The ductility is increased.
Hardness levels are reduced.
Toughness is slightly reduced at short
times but the effect can be significant over
longer times.
Effect on Creep Properties
For creep resisting material, post weld heat
treatment is required in order to fully
develop the creep strength. This is especially
true for thicker components such as headers.
There has been a tendency in recent years to
allow waiving of the post weld heat
treatment stage for thinner materials used
typically for superheater and reheater coils
in the Power Generation industry, but a
variety of conditions have to be met.
Other benefits
Improving the diffusion of hydrogen out
of weld metal
Softening the heat affected zone and thus
improving toughness (although not weld
metal toughness)
Improving dimensional stability during
machining.
Improving ductility.
Improving the resistance to stress
corrosion cracking.
Reducing the effects of cold work.
WHEN TO POST WELD HEAT TREAT
Within pressure equipment standards, the
requirement for post weld heat treatment is
largely a function of the material and the
thickness. The material (in terms of alloy
content) and the thickness (in relation to the
quench effect) control the microstructure
that will be formed. Large section
thicknesses in alloy steels can result in
martensitic, pearlitic or bainitic structures,
depending on the cooling rate, and this is
usually controlled by the use of preheat. In
addition, the thicker the material that is
welded, the greater the amount of residual
stress that will be developed on cooling.
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 3 of 10
For typical carbon-manganese steels, the
thickness at which post weld heat treatment
becomes mandatory is consistent in the 32
38 mm range for most of the Codes in use in
Australia. The reason for each standard
choosing a specific thickness is not entirely
clear, but little has changed over the past 30
years. What is interesting though is that
experiments conducted in the mid-1970s
showed that fully restrained butt welds in
carbon-manganese steels could develop
residual stresses in excess of the yield stress
at a thickness of approximately 35 mm.
With alloy steels, the thickness at which post
weld heat treatment becomes mandatory is
much less. Typically, the range is 13 20
mm, and even below 13 mm, a series of
strict conditions have to be met before post
weld heat treatment can be waived. It is
clear therefore, that with alloy steels, the
removal of residual stress is not the only
consideration for the application of post
weld heat treatment.
Post weld heat treatment of structural steels
is almost unheard of. Even in the offshore
industry, the Nodes and K-joints on the
Jackets are no longer post weld heat treated.
When that industry was in its infancy, post
weld heat treatment was mandatory and was
applied, but very little CTOD information
was available then, and the materials in use
suffered laminar tearing. That has all
changed now, and very few joints require
any attention today. Similarly, the massive
machines in use in the mining industry are
not subjected to post weld heat treatment,
and it is not even addressed in the range of
structural welding standards, AS 1554 parts
1 to 6. Therefore, the topic is essentially
limited to the pressure equipment industry.
HOW TO PERFORM POST WELD
HEAT TREATMENT
General
Most of the requirements for post weld heat
treatment can be found in the fabrication
Standard to which the vessel is constructed.
In Australia, most fabrication standards now
refer to AS 4458 for manufacture, and all
details can be found there. The American
Standards still tend to retain their own
requirements, and ASME I, ASME VIII
Division 1 and ASME VIII Division 2 are
all different. Part 4 of the new EN 13445
deals with manufacture, and contains a lot of
detail on post weld heat treatment.
Fixed Furnace
This method is traditional and well known to
most people in the fabrication industry. A
fixed furnace usually consists of a
rectangular box made from heat resistant
materials in which are embedded electric
resistance elements. Doors open at each end
and a travelling bogie allows for loading and
unloading of the charge. Furnaces such as
these are often capable of heating to 1200
0
C
and can normalise and anneal as well as
stress relieve. Some furnaces are gas fired,
with two or even four nozzles at each end.
Fixed furnaces tend to be large and
expensive to operate. They often have fixed
thermocouples that measure the furnace
atmosphere temperature and not the
temperature of the article being heat treated.
This is usually satisfactory up to around
300
0
C, but beyond, thermocouples
physically attached to the article must take
over both temperature control and
temperature measurement. Furnaces such as
these must be equipped with correctly
calibrated temperature controllers/recorders
with at least 12 recording points.
Care must be taken when cooling after post
weld heat treatment. Most manufacturing
Codes specify a controlled rate of cooling
until a certain temperature is reached
(typically 300 400
0
C depending on the
thickness), so it is normal to control cool in
the furnace before opening the doors.
Temporary Furnace
These are custom-built around a vessel,
rather than transport a vessel to a fixed
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 4 of 10
furnace. The idea is to minimise the air
space between the vessel and furnace walls,
and they allow for faster heating and
cooling. The basic structure of the furnace
should be creep resisting piping (if the pipes
are to be continually re-used) with heat
resistant materials attached to them. Heating
can be through resistant heating mats placed
on a concrete floor or via gas burners placed
at each end.
In the case of gas burners, care must be
taken to avoid direct flame impingement on
the vessel.
Temperature control is again through a 12
point recorder/controller, but atmosphere
thermocouples are not generally used. The
same heating and cooling rate restrictions
apply as with fixed furnaces.
Internal Firing
Vessels of suitable dimensions and
arrangement of openings can be post weld
heat treated by gas firing through nozzles or
manways. Manways are large enough to
accommodate the gas burners, but care
needs to be exercised with the diameter and
position of nozzles and expert opinion
should be sought. Care must also be taken to
place deflector plates inside the vessel and
opposite the burner entry points to avoid
direct flame impingement on the shell. It is
not advisable to post weld heat treat vessels
that contain internals in this manner.
The outside of the vessel must be
completely encased in insulating material,
and again, at least a 12 point temperature
recorder is advisable.
Local Heating
Circumferential weld seams can be post
weld heat treated by heating a band around
the weld. Although not specifically stated,
such heating is essentially limited to
resistance or induction heating, mainly
because of the controls required on heated
band width, width of insulation and
temperature measurement requirements. The
use of gas ring burners and other methods
should be carefully documented, proposed to
the Client and accepted before being
contemplated for use.
Typical restrictions on the width of the
heated band in both AS 4458 and EN 13445
involve the use of a formula, ie. 5(Rt
s
)

where the weld is in the centre and:

R = inside radius of the shell, in mm
t
s
= nominal shell thickness, in mm
This means that with a vessel of say 2 metre
inside diameter and 50mm wall thickness,
the minimum width of the heated band
would be
5 x (1000 x 50)

= 1118 mm
or 560mm each side of the weld.
In addition to this, the width of the insulated
band is recommended to be 10 (Rt
s
)

, which
in the above example would be 2236mm.
This is to ensure that the temperature
gradient of the portion next to the heated
band is not harmful. There may be other
restrictions or different formulae involved,
and the relevant Code must be checked.
Partial Heat Treatment
There are occasions, for example with very
long vessels, when the entire vessel will not
fit into a fixed furnace. This has been
catered for in most Standards, and it is
permissible to post weld heat treat section of
the vessel first, then turn the vessel around
and heat treat the remaining section. As with
local heating, there are restrictions in this
case as well over the degree of overlap and
the longitudinal temperature gradient. AS
4458 requires an overlap of at least
1500mm, and care should be taken to ensure
that this section does not contain a weld
which may be post weld heat treated twice.
Generally, there are better ways to post weld
heat treat vessels than to use this partial
method.
Thermocouples
For the temperature ranges applicable in this
guidance note, ie. up to 750
0
C, the normal
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 5 of 10
chromel-alumel thermocouples perform in a
satisfactory manner. They can be purchased
either as insulated wires or as sheathed units.
The wire type are better in that they can be
attached directly to the equipment being heat
treated by capacitor discharge. The sheathed
types need to be physically attached, often
by welding a drilled nut onto the equipment
and bolting the sheath into the nut. There is
an argument that says this measures the
atmosphere temperature just above the nut
and not the metal temperature, but the basis
of the argument is tenuous at best.
The number and placement of
thermocouples is also governed to a certain
degree by Standard requirements. For
example, AS 4458 stipulates that the
maximum variation in temperature within
the component shall not exceed 140
0
C
within any 5 metre length. EN 13445
stipulates a maximum variation of 150
0
C
within a 4500mm length, but restricts this to
100
0
C when the temperature exceeds 500
0
C. Sufficient thermocouples therefore need
to be attached to demonstrate that this
requirement is met. Allowance should also
be made for breakage and thermocouple
malfunction, so it is normal to attach more
than just the minimum. For internally fired
vessels and locally heat treated weld seams,
for each thermocouple attached to the
outside, there should be a corresponding one
on the inside. There is also a requirement for
locally heat treated weld seams and partially
heat treated vessels to measure the
temperature at the edge of the heating band
to ensure it is at least half of the peak
temperature. For vessels where there are
differing material thicknesses,
thermocouples should be placed on all
thicknesses, irrespective of the 5 metre
length.
Heating/Cooling Rates
These are specified in most of the
construction Standards, and are reasonably
similar. For example, AS 4458 for carbon
manganese steel vessels limits heating and
cooling rates to 200
0
C/hour for thicknesses
up to 25mm, and 5000
0
C/hour divided by
the thickness for vessels over 25mm. There
is also a minimum limit of 50
0
C/hour for
very thick vessels. However, for
CrMoV and 2Cr1Mo alloys, the
heating rate is limited to half that of carbon
manganese steels. For comparison, EN
13445 allows 220
0
C/hour up to 25mm thick,
5500/thickness for 25 to 100mm and 55
0
C/hour above 100mm thick. There are no
restrictions on the alloy steels in EN 13445.
By the time the holding temperature has
been reached, the variation in temperature
between any two thermocouples should not
exceed the range specified for the material.
In AS 4458, the range is typically 40
0
C (eg.
C-Mn steel range is 580 620
0
C), and in EN
13445, it is 50 70
0
C. In most practical
situations, the variation can be limited to
much less than this.
The issues described here are vital to avoid
post weld heat treatment going wrong and
the result being a scrap vessel. It is
recommended to all fabricators to obtain a
post weld heat treatment procedure from a
heat treatment company, where all of the
above-mentioned restrictions are spelled out,
before starting.
Test Pieces
There are two main types of test piece that
accompany pressure vessels through the post
weld heat treatment cycle. In most countries,
they are the Production Test Plate (PTP) and
the Coupon Plate. The PTP contains a weld
seam, and is either a run-off plate from a
longitudinal weld, or a weld done in parallel
to represent a circumferential weld. The
Coupon Plate is usually an off-cut of parent
material, and is used to follow the vessel
through all heat treatment cycles, including
hot forming, normalising and post weld heat
treatment.
It is highly unlikely that such test pieces
would be needed where local post weld heat
treatment has been applied. Test pieces are
therefore limited to furnace heat treated
vessels. Most Standards require that the test
piece(s) accompany the vessel through the
actual heat treatment cycle, and will not
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 6 of 10
allow a simulated cycle (AS 4458 is an
exception).
For internally fired post weld heat treatment,
the test pieces must be placed inside the
vessel, but with furnace methods, they may
be placed inside or outside the vessel. Care
must be exercised with the location, since
the test piece must be subjected to the same
heating and cooling rate as the vessel. It is
also critical for each test piece to have a
separate thermocouple attached.
Charts and Certificates
No post weld heat treatment should ever be
performed without a chart generated by a
correctly calibrated recorder. The chart is
proof that the post weld heat treatment was
carried out correctly, and forms part of the
Manufacturer's Data Report. In many
countries, the Third Party Inspector is
required to witness the loading of the
furnace or the set-up in local methods, and
actually sign the chart. In Australia, this is
not required, but a senior member of the
Company performing the post weld heat
treatment should sign the chart.
A Certificate of post weld heat treatment is
also often issued, describing the equipment
and giving details of job number, etc. but
this should be complimentary to the chart,
not a replacement for it.
Non-destructive testing
EN 13445 requires that all non-destructive
testing be carried out after post weld heat
treatment. There is a modification that
allows for material that is not sensitive to
heat treatment cracking to be examined
before post weld heat treatment. These
materials include carbon manganese, carbon
molybdenum and 1CrMo steels, but the
customer should be consulted before
assuming non-destructive testing may be
done before post weld heat treatment.
Of the American Codes, ASME B31.3 is
probably the most widely used in Australia.
It is slightly more restrictive than its
European counterparts in that only carbon
manganese steels can be tested before heat
treatment, and all alloy steels must be tested
afterwards. In this regard, ASME 1, VIII
Div.1 and VIII Div.2 are all similar to
B31.3.
AS 4458 refers to AS 4037, and this again
allows testing before or after post weld heat
treatment depending on the alloy.
Restrictions to testing after heat treatment
apply mainly to high alloy steels such as
martensitic, austenitic and ferritic types, as
well as the CrMoV steels. There is a
further restriction for 9%nickel steels and
quenched and tempered alloy steels in that
there has to be a 7 day delay between post
weld heat treatment and magnetic particle
testing, and that this testing must also be
done after hydrostatic testing.
It is strongly recommended that fabricators
fully understand the requirements of the
contract and the particular requirements of
the Standard before establishing their
inspection and test plans.
OMISSION OF POST WELD HEAT
TREATMENT
There are a number of situations where
pressure equipment requires work to be done
due to the service environment, and such
work often involves welding. Typical
situations include:
Repairs due to mechanical damage
Repairs due to a corrosion mechanism,
including cracking.
Repairs due to creep damage
Repairs to service-propagated defects.
Repairs to original manufacturing defects
Modifications to take advantage of the
economic situation.
Modifications due to changes in raw
material feed.
In some States in Australia, the use of AS
3788 is mandatory, and in the others, it is
used extensively as a guideline. This
standard, in paragraph 6.2.1 makes a clear
statement that all repairs shall be carried out
in accordance with the relevant design and
fabrication standards. It goes on to state in
paragraph 6.2.2 that minor repairs may be
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 7 of 10
carried out only for those materials and
service conditions, or both, which do not
require any post fabrication heat treatment.
Standards in other countries, eg. the
National Board Inspection Code in the
United States, make similar statements. At
first glance therefore, any repair, alteration
or modification to an item of pressurised
equipment that was originally post weld heat
treated after fabrication needs to be post
weld heat treated again after repair. This is
not always possible, and efforts over the past
20 years have been aimed at finding ways
around this.
Cost Advantages
Whilst most repairs that are contemplated
could be post weld heat treated, even if it
means removing the equipment to a
workshop, the cost of doing so becomes
prohibitive in todays economic climate.
There are other situations, notably in the
Nuclear Power industry, where removal of
equipment is simply not possible. Therefore,
in terms of risk management, there has been
a strong drive to develop techniques to
eliminate the need for post weld heat
treatment following an on-site repair.
Technical Advantages
This is quite an emotive issue, and like
statistics, any number of conclusions can be
drawn from sets of data. Of the effects that
post weld heat treatment has on a weld,
residual stress will obviously remain if post
weld heat treatment is waived, and for creep
resisting materials, the full creep strength
will not be developed. However, there are
welding techniques that can simulate the
tempering effect of post weld heat treatment,
and there are some claims that mechanical
properties in the HAZ can be improved
compared with conventional post weld heat
treatment. The particular property that is
influenced more than any other when
welding with these alternative techniques is
heat affected zone toughness. Most of the
so-called temper bead techniques are
primarily designed to give adequate
toughness in both the weld and the heat
affected zone and to produce a satisfactory
hardness profile. Whether the room
temperature, and sometimes sub-zero impact
properties are the most important
consideration with alloy steels operating in
the creep range is indeed questionable.
After having said that, there is no doubt that
for materials susceptible to reheat cracking,
the absence of post weld heat treatment and
the absence of the coarse-grained heat
affected zone are definite advantages. Care
needs to be exercised however, and suffice it
to say that the justification for waiving post
weld heat treatment on technical grounds
should not be confused with economics.
TECHNIQUES FOR WELD REPAIR
WITHOUT POST WELD HEAT
TREATMENT
Attempts to achieve satisfactory results from
weld repairs where post weld heat treatment
has been omitted are not new. Much of the
original work carried out for the Nuclear
Industry began in the early 1980s, and Code
Case N-432 for a limited range of materials
under ASME III in the United States was
established in 1986. Since then, the various
techniques have been refined and adopted
into other standards such as ASME B 31.1,
B 31.3 and even ASME XI. Some of the
better known techniques will be examined in
more detail.
Half bead technique
This technique was developed using the
shielded metal arc (SMAW) process and
was essentially aimed at providing an
alternative to the use of post weld heat
treatment. The technique was originally
developed for use in the nuclear industry,
but has since become widely used for repairs
to piping, headers and turbine casings in
conventional power plant and other
industries. The SMAW technique usually
employs the use of increasing diameter
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 8 of 10
electrodes, starting with a 2.5 mm, then a 3.2
mm and finishing with a 4.0 mm.
The area to be repaired is preheated to a
temperature commensurate with the material
and thickness, and a buttering technique
used as a first layer with the 2.5 mm
electrode. The purpose is to produce a small,
shallow heat affected zone. The next step is
to remove approximately half of the
thickness of each run by grinding. The
second layer deposited using 3.2 mm
electrodes effectively re-transforms the
coarse-grained heat affected zone and first
layer, and the third pass with the 4.0 mm
electrode further tempers the heat affected
zone. Each subsequent layer transforms and
tempers the layers beneath in the normal
way.
The major advantage of the technique is that
the toughness of the heat affected zone is
considerably improved over conventional
methods, but the disadvantages include the
fact that a lot of grinding is required, and
accurate grinding at that. Not only is this
time consuming, but if too much material is
removed from the first layer, the effects of
the re- transformation may not be as
successful as might be expected.
Consistent layer temper bead technique
This technique utilises either the SMAW or
the GTAW process, and was developed by
EPRI in the early 1990s to ensure that
toughness properties in both the heat
affected zone and the weld metal were at
least equal to the toughness properties of the
original base material. The technique
involves depositing weld layers that are
sufficiently thick that the subsequent weld
layer only tempers the heat affected zone
caused by the first layer. The temperature is
not intended to exceed the A
C1
so no grain
refinement occurs, and the effect is
essentially similar to a post weld heat
treatment. The technique produces a heat
affected zone microstructure that consists
predominantly of tempered martensite with
small amounts of bainite, resulting in good
toughness properties.
Alternate temper bead technique.
This technique was also developed under the
EPRI program but specifically for carbon-
manganese and carbon-molybdenum
materials used in nuclear reactor pressure
vessel components. It utilised the automatic
GTAW process and was an alternative to the
half bead technique for use in areas of high
radiation exposure. The technique involves
preparing the area to be repaired so that at
least six buttering layers can be performed.
A preheat of 150
0
C minimum is applied,
and the heat input of each layer is controlled
to within 10% of that measured in the
procedure qualification test. If more than six
layers are used, the same control over heat
input is required. The final step is a low
temperature post weld heat treatment at 230
290
0
C.
The aim is to produce tempering using
several weld layers so that each subsequent
layer penetrates the previous layer to
develop overlapping temperature profiles.
However, the A
C1
is exceeded, resulting in
re-transformation and grain refinement of
the heat affected zones. Sometimes,
different heat inputs are used for the first
three layers, starting with a low heat input to
minimise the extent of the heat affected zone
in the parent material. The heat input for the
second layer is increased to give a slightly
thicker deposit whilst still re-transforming
the parent heat affected zone, and the third
layer is still thicker to ensure tempering only
of the heat affected zone.
Controlled deposition technique.
This technique resulted from special cases
where creep embrittlement and reheat
cracking were potential problems during
repair, and was aimed at specific materials
used in conventional fossil-fired stations. It
is also a SMAW technique, and uses strictly
controlled ratios of heat input between one
weld layer and the next. The heat input for
the second layer is 1.3 to 1.8 times higher
than for the first layer, and is designed to
produce grain refinement and tempering in
the original heat affected zone. The ratios
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 9 of 10
need to be experimentally verified for each
material to be welded. The technique
eliminates the coarse grained region within
the heat affected zone where creep cracking
and reheat cracking have caused problems in
the past.
DECISIONS ON REPAIRS
Most decisions regarding repairs to
pressurised equipment in large, complex
plant are made using risk management
principles. On the one side of the risk
management equation is consequences of
failure, and this can be financial,
environmental or health and safety related.
Once an item of equipment has been found
to contain a flaw of some description, the
consequences of
Doing nothing,
Repairing the flaw,
Replacing the item,
need to be considered. If the decision is to
repair, the consequences of the repair
method also need to be evaluated. The
decision to repair must be made by the
Owner/Operator, who should also take full
responsibility for the consequences of the
decision. Decisions on the repair method are
more difficult. This is because in todays
climate, plant maintenance is often
subcontracted, and the subcontractor does
not always have the knowledge to advise on
repair consequences. The issues here tend to
be as much legal as technical, but again, the
Owner/Operator must take responsibility for
ensuring the correct advice is sought. In
certain industries, there are sufficient
Institutions (eg EPRI and UKAEA) and
people within those Institutions with the
knowledge and experience to advise on the
likelihood of success. In other industries,
this is not the case, and more is said of this
issue below.
On the probability of failure side and the
attendant risks, repairs where post weld heat
treatment has been omitted is reasonably
mature technology. In the power generation
and nuclear industries, there is sufficient
evidence to show that carbon-manganese,
carbon-molybdenum and chromium-
molybdenum steels operating at high
temperatures can be safely repaired. One
outstanding problem is the question of
properties when a defect is found in a
temper bead repair and a fracture mechanics
assessment has to be carried out. A review
of the literature shows that very little K
1C
testing has been carried out. The charpy
impact testing that has been done has been
done on the heat affected zone. If a defect is
in the weld repair, this immediately takes
any calculation into the realms of
probabilistic fracture mechanics.
In other industries such as Petrochemical,
the technology is probably less mature.
There are some instances where the temper
bead technique has been used quite
successfully in non-sour service. Whenever
NACE MR 0175 restrictions apply, repairs
without post weld heat treatment have
almost universally resulted in cracking of
the repair at a later date. This is quite
obviously the result of residual stress.
QUALIFICATION OF WELDING
PROCEDURES
In just about every world Code or Standard,
the application of post weld heat treatment is
an essential variable. This means that the
application of post weld heat treatment to a
procedure or the removal of it from a
procedure requires re-qualification. This is
because post weld heat treatment affects the
mechanical properties of the weld, and this
is the whole purpose of procedure testing.
There are even more restrictions if heat
treatment is above the lower transformation
temperature, but this guidance note has been
limited to the stress relieving aspect of post
weld heat treatment.
Caution must also be exercised when
working with the ASME code. The system
of supplementary essential variables can be
confusing, but basically, these are invoked
whenever notch toughness requirements are
WTIA-Panel 1 Guidance Note 6 PWHT February 2003 Disclaimer applies
Page 10 of 10
specified. In the case of post weld heat
treatment, there is an additional essential
variable that specifies that the procedure
qualification test shall be subjected to post
weld heat treatment essentially equivalent to
that to be encountered in production,
including at least 80% of the aggregate time
at temperature.
Taking a practical example of this, a 50mm
thick procedure qualification test plate
requiring impact testing and representing a
50mm shell plate would normally only have
been held at temperature for 2 hours (ie.
1hour/ 25mm). In practice, that same vessel
may have different shell thicknesses or even
a tube plate, and the 50mm shell may have
reached the holding temperature long before
the rest of the vessel. Quite often, the shell is
at temperature for 5 or 6 hours, so the
procedure qualification test needed to be at
temperature for 6 hours x 80%, ie 4 hours 50
minutes and not the 2 hours it received.
Under these conditions, re-qualification
would be necessary.
DISCLAIMER
While every effort has been made and all reasonable care taken to ensure the accuracy of the material contained
herein, the authors, editors and publishers of this publication shall not be held to be liable or responsible in any way
whatsoever and expressly disclaim any liability or responsibility for any injury or loss of life, any loss or damage costs
or expenses, howsoever incurred by any person whether the purchaser of this work or otherwise including but without
in any way limiting any loss or damage costs or expenses incurred as a result of or in connection with the reliance
whether whole or partial by any person as aforesaid upon any part of the contents of this Guidance Note. Should
expert assistance be required, the services of a competent professional person should be sought.